Rāmānujācārya seems to assume that the subject is a desiring agent. Is such a subject devoid of any further psychological organ, but the mind (manas), or does it stand in need of the complex ontology of psyche (as developed in Sāṇkhya)?
Strangely enough, the whole Tantrarahasya never mentions either buddhi (intellect) or ahaṅkāra ("I"). In Śālikanātha Miśra's Prakaraṇapañcikā, Tattvāloka (p.327 of Śāstri's critical edition), the "I" is lastly mentioned for the first time:
ahaṅkārāvalambanena punardehātmavādapratyavasthānam. naitadevam."ghaṭam ahaṃ jānāmīti" jñātur ahaṅkārāspadībhūtatvāt "ahaṃ gacchāmy ahaṃ sthūlaḥ" ityādāv ahaṅkārasya śarīra eva pravṛttyavivādāc charīram eva pratyakṣaṃ jñātṛ pratīyate, pratyakṣavirodhe cānumānam ātmānaṃ na labhata iti.
The passage follows the rejection of the Cārvāka view that consciousness arises out of the body, just like intoxication out of plants. My first translation:
[Siddhāntin]: and through the support of the concept of an "I" (ahaṅkāra), again there is a rebuttal of the theory that the self is the body.[Cārvāka opponent]: It is not so. In "I know the pot" the knower has the rank of the "I" (ahaṅkāra). Since it is so, there is consent of the usage of "I" in "I go", "I am fat" etc., only in regard to the body. Hence, the body alone is perceptible [and] it is seized as knower. And an inference which goes against direct perception does not hold.
Comments: I would rather expect ahaṅkārāspadībhūtatvāt and pravṛttyavivādāt to be on the same level: since the knower is equated to the "I" and since the "I" is found to be the same as the body, the body is the knower!
Philosophically speaking, the Cārvāka argument is a sort of syllogism:
1. every knower is an "I"
2. "I" is used for the body
3. HENCE, the knower is the body
What's wrong with that? Well, first that "I" risks to be used in two different senses (in the first case, as an agent, in the second, as referring to an outside perceptible entity, animated, but not necessarily identical with the first one). Hence, it sometimes presupposes the conclusion it wants to establish (the identity of the body with what "animates" it).
Secondly, the fact that a is b does not imply necessarily that b is a. For instance "All Athenians are Greek" does not imply "All Greek are Athenians". Similarly, it might be the case that the knower does not always coalesce with the "I". I would not subscribe with this view, I just think that –formally speaking– it is possible.
Can readers better acquainted with Cārvāka texts know more about the argument? Can one really argue that the body itself "knows"? How far is this position different from the one of Contemporary Neuroscientists arguing in favour of no further "subject" besides the brain?